Enteric Methane Emissions, Rumen Fermentation Characteristics, and Energetic Efficiency of Holstein Crossbred Bulls Fed Total Mixed Ration Silage with Cassava instead of Rice Straw

Abstract

This study investigated the effects of substituting rice straw with cassava in total mixed ration silage on feed intake, digestibility, rumen fermentation, ruminating activity, and energy balance. An energy balance experiment was conducted to determine nutrient and energy utilization in dairy bulls. Four Holstein Friesian (HF) crossbred young bulls (93.19% HF × 6.81% Native Thai; average age of 12.08 ± 0.22 months and body weight of 266 ± 9.80 kg) were used in a 4 × 4 Latin square design with four 28-d periods. The four dietary treatments included substituting cassava with rice straw on a 50, 150, 250, and 350 g/kg dry matter basis. Increasing the amount of cassava in the diet resulted in linearly decreased rumination behaviors and enteric methane emissions (p < 0.05) but a linear increase in feed intake and digestibility, total volatile fatty acid concentration, and propionic acid: acetic acid ratio in the rumen fluid (p < 0.05), thus leading to a greater energy balance status (p < 0.01). Our results indicated that replacing cassava with rice straw from 5% to 35% in the ration of Holstein bulls resulted in a substantial reduction in physically effective fiber and chewing time but improved nutrient and energy supply. The metabolizable energy requirement for the maintenance of crossbred dairy bulls was estimated to be 599.9 kJ/kg metabolic body weight and the efficiency of metabolizable energy used for growth was 0.88. We concluded that cassava is a good energy feed resource for Holstein crossbred bulls in the tropics.

1. Introduction

Ruminants are an important source of enteric methane that produce significant amounts of 30% of global human-induced methane emissions [1]. Enteric methane is a greenhouse gas synthesis from the rumen anaerobic fermentation process of methanogenic archaea and bacteria that causes a loss of enteric methane energy, accounting for 4.8% to 13.7% of gross energy feed intake [2,3]. Therefore, energy loss in ruminants through enteric methane emissions is a problem because of the impact on global warming and because it reduces the efficiency of feed energy utilization and livestock productivity [4,5].

Type of feed and feeding strategies to reduce enteric methane emissions are a priority in improving cattle productivity and environmental sustainability [2,4,5]. The primary reasons for the low productivity of ruminants in tropical developing country feeding systems are animal genetics, available feed resources, and feeding systems, which depend on low-quality forage and industrial byproducts that limit feed intake, digestibility, and energy supply [6,7,8]. Feed intake, nutrient digestibility, and the net energy supply of forage are influenced mainly by its content of digestible and indigestible neutral detergent fiber [6,7]. Compared with rice straw, cassava is a suitable feed resource for improved rumen digestibility because it contains high rumen fermentable carbohydrates with similar crude protein and fat contents [8,9]. Current reports suggest that increases in the energy content of diets using cassava instead of rice straw improve beef cattle’s feed intake, digestibility, and growth performance [8,9,10]. However, overfeeding rumen fermentable starch which is low in physically effective fiber to increase intake and meet the energy requirement of high-yielding cattle has the challenge of increasing the risk of subacute ruminal acidosis (SARA). Additionally, cassava pulp is a cassava starch factory byproduct containing more moisture and fiber when compared with cassava chips that are produced from cassava tubes (Cassava, Manihot esculenta Crantz) after chopping and sundry feedstuffs [8,9].

Fermented feed technology combines the benefits of a nutrient-balanced diet and ensilaged. High-moisture feeds, such as cassava pulp and brewer’s grains, are blended with dry feeds, such as rice straw [8,10,11]. Ensiled total mixed rations have superior storage time [8,11], aerobic stability [11], enhanced rumen fermentation characteristics [9], mitigated enteric methane [9], and improved growth performance of beef cattle [8,9,10].

However, more information is needed on the effects of cassava and rice straw in fermented total mixed ration feeding systems. Therefore, we aim to investigate the effects of substituting cassava with rice straw in a fermented total mixed ration on feed intake and digestibility, rumen fermentation, eating activity, and energy balance of Holstein young bulls.

2. Materials and Methods

The experiment was conducted at the Khon Kaen University research farm in Khon Kaen Province, Thailand (latitude 16.47° N, longitude 102.81° E) from December 2019 to March 2020. All procedures involving live animals were approved by the Animal Care and Use Committee of Khon Kaen University (Record No. IACC-KKU-49/62, Reference No. 660201.1.11/48).

2.1. Animals, Design, and Dietary Treatments

Four Holstein Friesian crossbred young bulls (93.19% HF × 6.81% Thai native (Zebu); average age of 12.08 ± 0.22 months and initial body weight (BW) of 266 ± 9.80 kg) were used in the energy balance trial experiment. The animals were housed individually in pens (2.5 × 4.5 m) equipped with rubber mats with free access to concrete feed bunks and automatic drinking water ad libitum throughout the experiment. Each animal was treated for internal and external parasites (1 mL/50 kg BW; Ivermectin and Clorsulon, Ivermectin F, Bangkok, Thailand) and vitamins A, D3, and E (5 mL/head; Phenix, Bangkok, Thailand).

The experimental design was a 4 × 4 Latin square in which each animal was randomly assigned to 1 of 4 dietary treatments alternating over four experimental periods; each period lasted 28 d, with the first 23 d for diet adaptation followed by 5 d of urine, fecal, and respiratory gas collections. The four dietary treatments included substituting cassava with rice straws on a 50, 150, 250, and 350 g/kg dry matter basis. The ingredients and chemical composition of the diets are shown in

Table 1. Analyzed chemical composition and feed cost of rice straw, cassava pulp, and cassava chips.

Item 1	Feed Ingredients
	Rice Straw	Cassava Pulp	Cassava Chip
Chemical composition, % of DM 1
Dry matter	91.90	19.47	88.85
Organic matter	85.77	96.80	96.68
Crude protein	4.10	2.68	2.50
Ether extract	0.89	0.40	0.79
Non-fiber carbohydrates	43.46	74.56	89.70
Acid detergent fiber	43.25	25.15	5.27
Acid detergent lignin	-	4.40	0.96
Neutral detergent fiber (NDF)	68.50	38.90	13.75
Forage NDF	-	-	-
pdNDF240	26.80	34.75	9.89
iNDF240	41.70	4.09	3.89
Calcium 2	0.38	0.37	0.16
Phosphorus 2	0.09	0.03	0.08
ME (Mcal/kgDM) 2	1.60	2.89	3.27
Feed cost (Thai baht/kgDM)	2.50	2.05	7.00

¹ pdNDF = potentially digestible NDF; iNDF = indigestible NDF; ME = metabolizable energy content ² Calculated value.

and

Table 2. Feed ingredients, fermentation profile, and analyzed chemical composition of the four fermented total mixed ration dietary treatments.

Item 1	Levels of Rice Straw in Diet (g/kg DM)
	50	150	200	300
Feed ingredients, g/kg DM basis
Rice straw	50	150	250	350
Cassava chip	200	150	100	50
Cassava pulp	350	300	250	200
Rice bran	55	55	55	55
Palm kernel cake	60	60	60	60
Soybean meal	140	140	140	140
Brewer’s grain	130	130	130	130
Urea	5	5	5	5
Mineral	5	5	5	5
Premix	5	5	5	5
Total	1000	1000	1000	1000
Fermentation profile, % of DM
pH	3.74	3.83	3.94	4.01
Lactic acid	13.15	17.09	12.80	15.33
Acetic acid	3.11	2.45	2.88	2.42
Propionic acid	0.29	0.26	0.22	0.23
Butyric acid	2.70	1.88	1.50	1.40
Chemical composition, % of DM
Dry matter	33.25	33.05	35.62	40.20
Organic matter	94.47	93.87	93.08	92.35
Crude protein	17.07	17.29	17.40	17.21
Ether extract	5.69	5.56	4.96	4.46
Non-fiber carbohydrates	31.16	27.51	25.61	24.51
Acid detergent fiber	21.76	23.91	23.79	32.23
Acid detergent lignin	3.71	3.76	3.88	3.98
Neutral detergent fiber (NDF)	41.91	43.82	45.69	46.40
Forage NDF	2.10	6.57	11.42	16.24
pdNDF240	30.02	26.81	24.63	20.66
iNDF240	11.89	17.01	21.06	25.74
Calcium 2	0.32	0.32	0.33	0.34
Phosphorus 2	0.34	0.34	0.35	0.35
ME (Mcal/kgDM) 2	2.64	2.53	2.42	2.31
Feed cost (Thai baht/kgDM)	7.61	6.92	6.23	5.54

¹ Minerals containing 93.72 g Ca /kg, 46.86 g P /kg, 107.78 g Na /kg, 18.56 g S /kg, 8.24 g Mn /kg, 7.49 g Zn /kg, 3.37 g Mg /kg, 1.17 g Cu /kg, 0.15 g Co /kg, 0.01 g K /kg, 0.04 g I /kg, and 0.02 g Se /kg (Mineral #0106410029, Dairy Farming Promotion Organization of Thailand, Saraburi, Thailand). Premix containing 5,000,000 IU vitamin A /kg, 1,000,000 IU vitamin D3 /kg, 10,000 IU vitamin E /kg, 25 g Fe /kg, 4 g Cu /kg, 20 g Mn /kg, 0.13 g Co /kg, 15 g Zn /kg, 0.75 g I /kg, 0.38 g Se /kg, 0.20 g/kg feed preservative, and 0.88 g/kg feed additive (Golden Mix S # 0104610040, DFC Advanced Co. Ltd., Khon Kaen, Thailand). pdNDF = potentially digestible NDF; iNDF = indigestible NDF; ME = metabolizable energy content. ^2.Calculated value.

. The formulated diets contained an isonitrogenous diet (approximately 17% crude protein) to meet the nutrient requirements of a growing bull weighing 300 kg of shrunk body weight and with a target average daily weight gain of 1.57 kg, according to the guidelines of the National Research Council (NRC) [12]. Animals were fed a diet twice daily at 800 and 1600 h. Animals were weighed at the beginning and end of each period to calculate energy intake and excretion based on metabolic body weight.

The fermented total mixed ration was prepared monthly using a horizontal mixer (Celikel TMR feed mixer, 108 Agriculture Machine and Equipment Co., Ltd., Lop Buri, Thailand). Approximately 2000 kg of dietary treatment mixtures were mixed and 35 kg per silo bag was packed. The silo bags were kept outdoors (25 °C to 36 °C) until use [8].

2.2. Feed Intake and Digestibility

Feed intake was measured individually and calculated from the difference between the offered and the refused feed. Each bull’s total collection of feed offered, feed refused, feces, and urine was weighed and recorded over the last five consecutive days of the experimental period. The apparent digestibility was determined by total fecal collection techniques, with feces excreted immediately collected daily into a pan, weighed, subsampled (10%, wt/wt), and stored at −20 °C. At the end of the collection period, 1 kg daily aliquot samples of feed offered, feed refused, and feces were thoroughly mixed and dried at 60 °C for 72 h in an oven and then ground (1 mm screen) before chemical analysis. The total urine volume was collected into plastic canisters containing 6 N HCl to maintain a pH below 3.0, weighed, aliquoted into samples (120 mL), and stored at −20 °C until analysis.

2.3. Feed Sampling and Chemical Analyses

The representative samples of dietary treatments, feed refusal, and feces were collected from each period to determine dry matter (DM; method 967.03), ash (method 942.05), ether extraction (EE; method 920.39), and crude protein (CP; method 984.13) according to AOAC [13]. Neutral detergent fiber (NDF) and acid detergent fiber (ADF) contents were analyzed using an ANKOM fiber analyzer (ANKOM 200/220, ANKON Technology, Macedon, NY, USA) [14,15]. Acid detergent lignin (ADL) content was analyzed through the solubilization of cellulose with 72% sulfuric acid. The non-fiber carbohydrate (NFC) content was calculated as follows: NFC (%DM) = 100 − (CP + NDF + EE + ash). The gross energy (GE) contents of the feed offered, feed refused, and feces and urine were determined using a bomb calorimeter (IKA C2000 Basic, IKA-Werke, Staufen, Germany).

2.4. Particle Size, peNDF, pdNDF, and iNDF Determination

The particle size was determined using the Penn State Particle Size Separator (NASO, Fort Atkinson, WI, USA). The physical effectiveness factor (pef₈) was defined as the proportion of DM retained on 8 and 19 mm sieves [16]. The physically effective NDF (peNDF₈) was calculated by multiplying the NDF concentration of the feed (% of DM) by the pef₈.

Preliminary tests for various feed ingredients were conducted to determine the potentially digestible NDF (pdNDF) and indigestible NDF (iNDF), and then to formulate the pdNDF and iNDF of the dietary treatment content [17,18]. Long-term in vitro fermentations (240 h incubation) were conducted according to the modified technique of Sommart et al. [19,20]. The digestibility of the remaining fiber was determined for pdNDF and iNDF as follows:

IVNDFD (%) = ((NDF initial incubation (g) − NDF after incubation (g)) × 100)/ NDF initial incubation (g)

(1)

pdNDF (%) = (NDF content (%) × IVNDFD (%))/100

(2)

iNDF (%) = NDF (%DM) − pdNDF (%DM)

(3)

2.5. Eating, Ruminating, and Chewing Behaviors

Eating and ruminating times were monitored visually for 24 h on day 23 of each period. Eating was defined as muzzling in or over the feed through with cattle chewing or swallowing. Ruminating was defined as regurgitation, chewing, and swallowing of a bolus. The cattle behaviors were recorded every 5 min and assumed to persist for the entire 5 min interval [8]. Total chewing time was calculated as the sum of total eating and ruminating. The total time spent resting was calculated as 24 h minus the total time spent chewing [8].

2.6. Animal Calorimetry and Energy Utilization Measurements

Four open-circuit respiration chambers (105 × 80 × 173 cm³) were used to measure the volume of oxygen consumption, methane production, and carbon dioxide production of each bull in each experimental period according to the methods of Suzuki et al. [21]. The oxygen concentration of the inlet and outlet from each bull was analyzed with an oxygen analyzer (4100 Gas Purity Analyzer, Servomex Group, East Sussex, UK). Enteric methane and carbon dioxide concentrations were measured with an infrared analyzer (IR200 Infrared Gas Analyzer, Yokogawa Electric Co., Tokyo, Japan). The airflow rate of approximately 450 L/min was recorded using a flow meter (Model NFHY-R-O-U, Nippon Flow Cell, Tokyo, Japan). Before starting each measurement, the gas analyzer was calibrated daily using 18.9% and 24.9% oxygen, 1.89% carbon dioxide, and 1960 ppm of methane standard gas (Linde (Thailand) Public Co., Ltd., Samutprakarn, Thailand). The recovery rate ranged from 98 to 104% for all chambers.

The metabolizable energy intake (MEI) was calculated from the difference between the digestible energy (DE) intake minus the urine and methane energy outputs. Heat production (HP) of the bulls was estimated according to the Brouwer method [22] as HP (MJ/d) = 16.18 × oxygen consumption (L/d) + 5.02 × carbon dioxide production (L/d) − 5.99 × urinary nitrogen excretion (g/d) − 2.77 × methane production (L/d). Energy retained (ER) was calculated as the difference between HP and MEI. The ME requirement for maintenance (ME_m) and the efficiency of energy utilization for growth (k_g) was estimated by dividing ER [23].

2.7. Ruminal Fermentation and Blood Metabolites

On day 19 of each experimental period, approximately 200 mL of rumen fluid was collected by rumen stomach tube techniques at 0 h and 3 h after morning feeding. The ruminal fluid was immediately measured for pH (Eutech pH 700, Eutech Instruments Pte Ltd., Ayer Rajah Crescent, Singapore). The ruminal fluid samples were filtered through three layers of gauze. One hundred milliliters of rumen fluid were mixed in a plastic test tube with 10 mL of 6 N HCl. Rumen fluid lactic acid and volatile fatty acid concentrations were determined using gas chromatography (GC2014, Shimadzu, Tokyo, Japan) [24]. The concentration of NH3-N was analyzed using a spectrophotometer (UV/VIS Spectrometer, PG Instruments, London, UK) [25].

When the rumen fluid was collected, a blood sample was also collected from each animal. Approximately 10 mL was taken from the jugular vein in a sterilized vacuum tube (Greiner Bio-One (Thailand) Ltd., Chonburi, Thailand), packed on ice, and transported to the laboratory for plasma analyses. Plasma urea nitrogen, glucose, triglyceride, cholesterol, total protein, and albumin concentrations were determined using colorimetric method test kits (Roche Diagnostics, Indianapolis, IN, USA) and an automated analyzer (COBAS INTEGRA 400 plus analyzer, Roche Diagnostics, IN, USA).

2.8. Statistical Analysis

All data were subjected to analysis of variance using the general linear model of SAS version 9.0 using the following model:

Y_ijk = μ + ρ_i + γ_j + τ_k+ ε_ijk,

where Y_ijk is a dependent variable, μ is the mean for all observations, ρ_i is the fixed effect of the period (i = 1 to 4), γ_j is the random effect of the cattle (j = 1 to 4), τ_k is the fixed effect of the diet (k = 1 to 4), and ε_ijk is the residual error.

The results are presented as the mean values and standard error of the means. Polynomial orthogonal contrasts were used to test linear, quadratic, and cubic responses. Means were compared using Duncan’s multiple range test. Significance was declared at p ≤ 0.05 and tendencies were noted if 0.05 < p ≤ 0.10.

3. Results

3.1. Chemical Composition of Diets

Feed ingredients and analyzed chemical composition are shown in Table 2. The CP content in the dietary treatment was similar among diets. The fiber content of NDF, iNDF, ADF, and ADL increased substantially when increasing the rice straw in the diets.

3.2. Feed Intake and Digestibility

Daily feed intake and digestibility are reported in

Table 3. Effects of replacing cassava with rice straw in the fermented total mixed ration of Holstein Friesian crossbred young bulls on feed intake and nutrient digestibility.

Item	Levels of Rice Straw in Diet (g/kg DM)				SEM 1	p-Value 2
	50	150	250	350		L	Q	C
Feed intake, kg/d
Dry matter	8.08 a	7.29 ab	6.55 b	6.44 b	0.40	0.03	0.97	0.60
Nutrient digestibility, %
Dry matter	78.71 a	69.88 b	61.96 c	54.29 d	1.70	<0.01	0.60	0.82
Organic matter	81.02 a	72.70 b	65.85 c	59.54 d	1.54	<0.01	0.48	0.93
Crude protein	78.59 a	69.14 b	67.14 c	61.79 c	2.17	<0.01	0.82	0.49
Ether extract	89.66 a	89.08 a	79.69 b	80.30 b	1.55	<0.01	0.88	0.09
Neutral detergent fiber	71.64 a	61.19 b	53.90 c	44.53 c	1.85	<0.01	0.03	0.56
Acid detergent fiber	65.69 a	55.91 b	44.42 c	46.76 c	2.34	<0.01	0.05	0.17

¹ SEM, standard error of the mean. ² Polynomial orthogonal contrast probability of a significant linear (L), quadratic (Q), and Cubic (C). ^a–d Mean value in the same row with different superscripts differ (p < 0.05).

. An increased level of rice straw in the diet resulted in a significant linear decrease in daily feed intake (p < 0.05) and total tract nutrient digestibility of DM, OM, CP, EE, NDF, and ADF (p < 0.01).

3.3. Particle Size and Physically Effective Fibers

The distribution of particle size as a percentage retained on the sieve was significantly different (p < 0.01) among the treatments (

Table 4. Effects of replacing cassava with rice straw in the fermented total mixed ration on the particle size distribution and physical properties.

Item 1	Levels of Rice Straw in Diet (g/kg DM)				SEM 2	p-Value 3
	50	150	250	350		L	Q	C
Particle size, % DM retained on the sieve
19 mm	12.28 c	14.11 bc	15.08 ab	18.04 a	0.94	<0.01	0.84	0.88
8 mm	44.95 a	42.64 b	42.69 b	40.36 c	0.49	<0.01	0.98	0.08
2.36 mm	34.21 a	33.61 a	31.30 ab	28.35 b	0.86	<0.01	0.22	0.79
1.18 mm	7.82 c	9.04 b	9.57 b	12.34 a	0.24	<0.01	0.01	0.03
Pan	0.75 ab	0.60 b	0.64 b	0.91 a	0.05	0.06	<0.01	0.88
Pef8	0.57	0.57	0.58	0.58	0.01	0.18	0.81	0.29
peNDF8,%	23.98 b	24.87 b	26.72 a	27.10 a	0.36	<0.01	0.50	0.18

¹ Pef₈ = physical effectiveness factor, calculated as the total DM retained on 8 and 19 mm sieves; peNDF₈ = physically effective NDF, determined as NDF concentration of TMR multiplied by physical effectiveness factor. ² SEM, standard error of the mean. ³ Polynomial orthogonal contrast probability of a significant linear (L), quadratic (Q), and Cubic (C). ^a–c Mean value in the same row with different superscripts differ (p < 0.05).

). No differences in the pef₈ occurred among treatments. The increasing rice straw level in the diets linearly increased the particles retained on the peNDF, 1.18, and 19 mm sieves (p < 0.01), whereas the particles retained on 8 mm and 2.36 mm sieves decreased linearly (p < 0.01).

3.4. Chewing Behaviors

The Holstein bulls’ eating, ruminating, chewing, and resting behaviors are shown in

Table 5. Effects of replacing cassava with rice straw in the fermented total mixed ration of Holstein Friesian crossbred bulls on eating, ruminating, chewing, and resting behavior.

Item 1	Levels of Rice Straw in Diet (g/kg DM)				SEM 2	p-Value 3
	50	150	250	350		L	Q	C
Eating, h/d	4.20	3.73	4.17	3.89	0.26	0.70	0.75	0.28
Ruminating
h/d	3.10 b	6.12 a	6.51 a	7.40 a	0.35	<0.01	0.08	0.06
min/kg DMI	26.47 c	48.92 b	54.95 ab	64.71 a	3.79	<0.01	0.16	0.41
min/kg NDFI	63.17 b	111.62 a	120.28 a	139.45 a	8.69	<0.01	0.12	0.37
min/kg pdNDFI	95.52 b	166.73 a	177.48 a	198.55 a	12.89	<0.01	0.11	0.39
min/kg iNDFI	224.97 b	349.80 ab	356.05 a	377.76 a	33.39	0.03	0.18	0.44
min/kg peNDFI	78.54 c	145.86 b	168.97 ab	197.58 a	11.42	<0.01	0.15	0.51
Chewing, h/d	7.29 c	9.85 b	10.68 ab	11.29 a	0.24	<0.01	0.03	0.07
Resting, h/d	16.71 a	14.15 b	13.32 bc	12.72 c	0.24	<0.01	0.03	0.07

¹ DMI = dry matter intake; NDFI = neutral detergent fiber intake; pdNDFI = potentially digestible NDF intake; iNDFI = indigestible NDF intake; peNDFI = physically effective NDF intake. ² SEM, standard error of the mean; BW, body weight; BW^0.75, metabolic body weight. ³ Polynomial orthogonal contrast probability of a significant linear (L), quadratic (Q), and Cubic (C). ^a–c Mean value in the same row with different superscripts differ (p < 0.05).

. Eating times were not significantly different among treatments. Cattle fed with the highest rice straw level in the diet took the longest time for ruminating (7.40 h/d), while cattle fed with the lowest rice straw level spent the shortest time (3.10 h/d). Increasing rice straw levels in the diets significantly linearly (p < 0.01) increased ruminating time. Total chewing time increased linearly (p < 0.01) when the rice straw level increased in the diet. Resting time decreased linearly (p < 0.01) when rice straw was increased in the diet.

3.5. Ruminal Fermentation and Plasma Metabolites

Ruminal fermentation characteristics and plasma metabolites are reported in

Table 6. Effects of replacing cassava with rice straw in the fermented total mixed ration of Holstein Friesian crossbred bulls on ruminal fermentation characteristics and plasma metabolites.

Item	Levels of Rice Straw in Diet (g/kg DM)				SEM 1	p-Value 2
	50	150	250	350		L	Q	C
Ruminal fermentation characteristics
pH	6.79 b	6.94 ab	6.99 ab	7.13 a	0.07	0.06	0.85	0.26
NH3, mg/dL	5.41	6.56	4.56	5.26	0.75	0.49	0.77	0.13
Lactic acid, mmol/L	0.35 a	0.27 ab	0.24 ab	0.20 b	0.04	0.04	0.64	0.87
Total volatile fatty acid, mmol/L	85.48 a	76.99 ab	71.24 ab	65.01 b	4.14	0.04	0.13	0.15
Acetic acid, mol/100 mol	63.84 c	65.60 b	66.85 b	70.09 a	0.50	<0.01	0.19	0.31
Propionic acid, mol/100 mol	22.97 a	21.24 ab	19.81 bc	17.98 c	0.67	<0.01	0.94	0.82
Iso-butyric acid, mol/100 mol	1.41	1.52	1.62	1.64	0.08	0.08	0.61	0.85
Butyric acid, mol/100 mol	8.14	8.64	9.17	8.06	0.60	0.92	0.23	0.55
Iso-valeric acid, mol/100 mol	2.27 a	1.79 ab	1.66 ab	1.49 b	0.20	0.03	0.46	0.69
Valeric acid, mol/100 mol	1.37 a	1.21 ab	0.90 bc	0.74 c	0.10	<0.01	1.00	0.49
Acetic: Propionic acid ratio	2.83 c	3.15 bc	3.38 b	3.95 a	0.13	<0.01	0.36	0.45
Plasma metabolites
Glucose (mg/dL)	88.5	85.0	86.8	83.8	1.84	0.18	0.90	0.27
Urea nitrogen (mg/dL)	14.0	14.8	13.5	13.5	1.20	0.51	0.49	0.29
Total protein (mg/dL)	6.5	6.3	6.4	6.3	0.15	0.43	0.81	0.47
Cholesterol (mg/dL)	90.0	92.8	93.5	105.0	6.80	0.17	0.52	0.68
Triglyceride (mg/dL)	16.3	19.5	16.3	23.3	1.88	0.08	0.36	0.09

¹ SEM, standard error of the mean. ² Polynomial orthogonal contrast probability of a significant linear (L), quadratic (Q), and Cubic (C). ^a–c Mean value in the same row with different superscripts differ (p < 0.05).

. Among dietary treatments, blood glucose, urea nitrogen, total protein, cholesterol, and triglyceride concentrations did not significantly differ (p > 0.05). No differences in ruminal ammonia nitrogen concentration occurred among treatments. The lactic acid, total VFA, propionic acid, iso-valeric acid, and valeric acid concentrations decreased linearly (p < 0.05). Still, ruminal pH increased linearly (p < 0.05) as the rice straw level increased in the diet, with acetic acid concentration and acetic acid: propionic acid ratio increasing (p < 0.01), whereas iso-butyric acid and butyric acid were not affected by diet treatment (p > 0.05).

3.6. Respiratory Gas

Respiratory gas data are presented in

Table 7. Effects of replacing cassava with rice straw in the fermented total mixed ration of Holstein Friesian crossbred bulls on daily respiratory gas (oxygen consumption, carbon dioxide production, and enteric methane emission).

Item 1	Levels of Rice Straw in Diet (g/kg DM)				SEM 2	p-Value 3
	50	150	250	350		L	Q	C
Oxygen, L/d	2919.6	2750.9	2709.8	2694.4	172.48	0.56	0.80	0.35
Carbon dioxide, L/d	3475.7 a	3371.7 ab	3204.6 bc	3119.5 c	42.97	0.04	0.03	0.03
Enteric methane
L/d	146.15 b	167.85 ab	200.23 ab	222.94 a	5.23	0.05	0.54	0.49
L/kg DMI	22.06	24.62	28.26	29.10	2.48	0.08	0.34	0.26
MJ/d	5.78 b	6.64 ab	7.92 ab	8.81 a	0.21	0.05	0.54	0.49
MJ/MJ of GEI (%)	4.96 b	5.69 ab	6.67a	6.72 a	0.50	0.04	0.29	0.22

¹ DMI = dry matter intake. ² SEM, standard error of the mean. ³ Polynomial orthogonal contrast probability of a significant linear (L), quadratic (Q), and Cubic (C). ^a–c Mean value in the same row with different superscripts differ (p < 0.05).

. Daily oxygen consumption was not affected by the diets, whereas carbon dioxide production decreased quadratically (p < 0.03) when the rice straw level increased in the diet. An increasing rice straw level in the diet resulted in a significant linear increase (p < 0.05) in enteric methane emissions (L/d, MJ/d, and %MJ/GEI).

3.7. Energy Partitioning and Efficiency of Metabolizable Energy Utilization

The energy intake and partitioning are shown in

Table 8. Effects of replacing cassava with rice straw in the fermented total mixed ration of Holstein Friesian crossbred bulls on daily energy intake, energy partition, and energy efficiency.

Item 1	Levels of Rice Straw in Diet (g/kg DM)				SEM 2	p-Value 3
	50	150	250	350		L	Q	C
Energy intake, kJ/kgBW0.75
GE intake	1604.00	1513.90	1483.70	1408.60	150.84	0.29	0.66	0.70
DE intake	1292.50 a	1063.30 ab	933.70 ab	797.80 b	124.70	0.03	0.51	0.77
ME intake	1145.60 a	916.10 ab	778.30 ab	654.40 b	113.91	0.02	0.46	0.74
Energy partition, kJ/kgBW0.75
Feces excretion	321.81 c	450.58 b	549.94 ab	610.80 a	31.70	<0.01	0.66	0.57
Urine excretion	67.81	64.90	55.75	52.34	9.77	0.35	0.89	0.48
Methane emission	67.02 b	82.50 ab	97.20 ab	103.49 a	2.84	0.05	0.50	0.39
Heat production	729.21	696.21	737.87	683.71	32.03	0.57	0.77	0.35
Energy balance	416.34 a	256.59 b	66.67 c	–97.82 d	14.68	<0.01	0.25	0.47
Energy efficiency
DE/GE	0.81 a	0.70 b	0.63 bc	0.56 c	0.02	<0.01	0.35	0.88
ME/GE	0.72 a	0.60 b	0.53 bc	0.46 c	0.03	<0.01	0.31	0.89
ME/DE	0.89 a	0.86 ab	0.83 bc	0.82 c	0.01	<0.01	0.37	0.55

¹ GE = gross energy; DE = digestible energy; ME = metabolizable energy; EB = energy balance. ² SEM, standard error of the mean. ³ Polynomial orthogonal contrast probability of a significant linear (L), quadratic (Q), and Cubic (C). ^a–d Mean value in the same row with different superscripts differ (p < 0.05).

. Gross energy intake was not different between treatments (p > 0.05). Digestible and metabolizable energy intake decreased linearly with increasing rice straw levels in the diets (p < 0.05). Feces energy and methane energy loss increased significantly (p < 0.01) when increasing the rice straw levels in the diets, whereas urine energy loss and total heat production were not affected (p > 0.05). As the rice straw level increased, there was a significant linear increase in the negative energy balance status (p < 0.01). There was a significant linear decrease (p < 0.01) in energy efficiency when increasing the rice straw level in the diet.

The metabolizable energy requirement for maintenance (ME_m) of growing Holstein crossbred bulls (Figure 1), as derived from the regression equation, was estimated at 599.99 kJ/kg BW^0.75. In addition, the efficiency of ME utilization for growth (k_g) was 0.88.

4. Discussion

This study examined the effects of replacing cassava with rice straw in a Holstein crossbred bull diet on feed intake and digestibility, rumen fermentation, eating activity, and energy partitioning. When cassava replaced rice straw in the diets, overall nutrient and energy intake, rumen fermentation end-products, reduced enteric methane emissions, and positive energy balance improved in Holstein bulls fed a fermented total mixed ration.

4.1. Chemical Composition of Fermented Total Mixed Ration

Fermented feed and feeding technology have the advantages of improving the nutrient balance, long storage time, and aerobic stability of agricultural byproducts that have high moisture. The fermented feed should have an elevated lactic acid content, but low pH and trace volatile fatty acid and be free of mycotoxin contamination [8,10]. This study prepared the fermented feed by mixing wet byproducts, such as cassava pulp and brewer’s grains, with dry feed ingredients, such as rice straw and oilseed cake; preserving them as silage produced an excellent fermentation characteristic that was well preserved, including a lower pH than <4.2 within 7 d of ensiling. It effectively maintained nutritive and economic values and remained well preserved for over three months. This finding was consistent with previous reports [8,9,10].

In the present study, the chemical composition and nutritive values of rice straw were similar in crude protein content, but differed in fiber fraction; therefore, the greater fiber of NDF, ADF, and ADL concentrations of rice straw compared with cassava resulted in more fibrous diets when cassava was replaced by rice straw. Diets containing rice straw also had a greater proportion of long particle size (retained on the >8 mm sieve), peNDF, and iNDF, indicating that rice straw was a superior source of physically effective fiber compared with cassava. These results agreed with those which previously reported that the proportion of particles on 19 mm and 8 mm sieves increased with increasing proportions of forage, such as rice straw, in the treatment diet [26,27]. Our data indicated that an increase in the NDF level in the diet was associated with a linear increase in the peNDF₈ (range 23.9 to 27.1%), providing sufficient peNDF requirements for Holstein young bulls. The minimum recommended 21% peNDF was required to sufficiently stimulate chewing activity and maintain an average ruminal pH greater than 6.0 above SARA risk [27]. However, rice straw also had higher ADL and iNDF than cassava, and less pdNDF and non-fiber carbohydrates, thus resulting in decreased feed intake and total-tract digestibility of nutrients when rice straw replaced cassava in the diets.

4.2. Feed Intake and Digestibility

Nutrient intake and digestibility limit the energy supply required for maintenance and production. Tropical feedstuffs generally contain a high lignocellulose fraction and bulky and slow-digesting components. High rumen degradation of organic matter, especially of non-fiber carbohydrate and fiber fractions, is the critical factor in increasing the dry matter intake of ruminants [8]. One explanation is that when fiber digestibility improves, non-fiber carbohydrates are digested faster in the rumen, allowing cattle to consume more feed [28]. The degradability of the dietary fiber fraction from the rumen is also an essential factor limiting feed intake and energy supply, as reported by Chaokaur et al. [4]. Raffrenato et al. [7] suggested that the gut-filling effect of the diet could be related to the content and rate of degradation of pdNDF. Raffrenato et al. reported that dry matter intake was inversely associated with iNDF [7]. Other studies have similarly observed a decrease in dry matter intake with low digestibility of NDF in diets [8,28,29,30,31,32]. Our results indicate that the rice straw level in the fermented total mixed ration diet substantially reduced feed intake and digestibility because of its indigestible fiber concentration and rumen distension potential filling effect in growing cattle fed tropical feedstuff diets.

4.3. Eating and Ruminating Activity

Ruminating activity is influenced by the cell wall content of diets, resulting in increased time spent ruminating. The ruminating time in this study ranged from 3.10 to 7.40 h/d. The linear increase in total chewing and ruminating time with the increasing proportion of rice straw is associated with more significant forage NDF, peNDF, and iNDF concentration. Total resting time decreased as the iNDF level increased in the diet. In our experiment, cattle fed the highest rice straw level consumed greater iNDF levels in the diet and took the longest time to ruminate (7.40 h/d), while cattle fed the lowest iNDF level in the diet spent the shortest time (3.10 h/d). These results are consistent with previous reports [32,33,34,35] which suggested that increased chewing activity was only caused by increased rumination activity. In addition, previous reports found that there are relationships between the NDF content and rumination time for dairy heifers [36]. The increased NDF intake and rumination time associated with increasing NDF content has been reported recently [32,37,38,39]. This finding suggests that the intake of forage NDF, the peNDF, and iNDF levels are the significant drivers for chewing and ruminating activity, thus promoting rumen function and animal health.

4.4. Ruminal Fermentation Characteristics and Enteric Methane Emission

Rumen microbes provide nutrients and energy sources to host animals by degrading and fermenting organic compounds from feed into cell biomass and rumen fermentation end-products such as amino acids, vitamins, minerals, and short-chain fatty acids. It was expected that cattle fed rice straw instead of cassava in the diet would have less total volatile fatty acids [20]. In this study, ruminal pH (ranging from 6.79 to 7.13) was relatively higher than 5.2 to 6.0 above the risk level of SARA, indicating no SARA susceptibility via the effective rumination and saliva production stimulated by the long particle size of rice straw in the diets. There was no sign of SARA illness, including a decrease in dry matter intake, laminitis, loss of body condition, and alterations in fecal consistency and structure. This result was in good agreement with other reports [8,20,33,35]. Our data also indicated that in the shift in the rumen fermentation process, the total volatile fatty acid concentration was associated with increased propionate and decreased acetate concentration. Propionate can provide energy for the cows by converting to blood glucose in the liver and converting in the mitochondria to produce citrate that contributes later to acetyl-CoA by ATP-citrate lyase in the cytoplasm as a precursor for fatty acid synthesis. Therefore, increased milk fat production or intramuscular fat deposition via increasing propionate may occur when lactating cows have a negative energy balance status with sufficient acetate or dietary fatty acids supply. Rumen fibrolytic bacteria that digest structural carbohydrates (hemicellulose and cellulose) produce a large proportion of acetic acid to propionate ratio, carbon dioxide, and hydrogen [20,40,41]. The increasing level of pdNDF and iNDF in the diet elevated the substrate for rumen archaea and methanogenic bacteria to produce methane, thus explaining the increasing enteric methane energy loss. Our previous studies have also suggested that a high-fiber diet results in greater dietary energy loss as methane through the effects on shifting rumen microbial populations and an increase in the acetate: propionate ratio [2,8,9,20].

In this study, the proportion of acetic acid and acetic acid to propionic acid ratio increased linearly in response to increased dietary NDF, agreeing with previously reported findings [20,42]. These can be expected to be related to higher methane production due to changes in the hydrogen balance, such that high acetic acid and butyric acid production enhance methane production. Our results demonstrated that increasing rice straw in the diet increased total fiber and indigestible fiber intake, thus remarkably increasing enteric methane emission (L/d) and methane energy loss (MJ/d). This result reflected the findings of other reports [2,10]. An increased intake of iNDF may stimulate the rumen fermentation pathways that promote acetic acid, hydrogen, and carbon dioxide production, thus the methanogen hydrogen and carbon dioxide precursor to produce methane [43].

In the present study, increasing cassava levels in the diets resulted in a significant linear reduction in enteric methane emission and energy loss. The enteric methane conversion factor is critical in estimating greenhouse gas emissions and the global warming impact in national inventories. In this study, the enteric methane conversion factor ranged from 4.9 to 6.7% MJ methane/MJ of gross energy intake (Ym), which is within the IPCC 7.0% default value for cattle fed low-quality crop residues and byproducts in developing countries [44]. Thai native and Charolais crossbred cattle have reported no differences in the methane conversion factor (ranging from 5.7 to 6.3% Ym) [14]. Our results also agree with previous works [8,21] reporting a relatively high rate of enteric methane conversion in cattle fed high NDF fiber diets. The typically high indigestible fiber feed in tropical feeding systems may mainly affect the enteric methane conversion factor. Improving digestible fiber forage to increase intake above maintenance can be an essential strategy to reduce enteric methane emissions [43]. These data highlight the feeding systems’ importance in improving animal production performance and environmental sustainability in the tropics.

4.5. Energy Partitioning and Energy Requirement for Maintenance

Our data demonstrated that decreasing rice straw levels in the diet improves daily feed and nutrient intake, resulting in an increased energy supply in Holstein bulls. Increases in metabolizable energy intake were positively related to decreases in energy loss from feces and enteric methane; therefore, a greater positive energy balance was achieved when cassava diets replaced rice straw diets. The positive energy balance is consistent with a previous report [4,8,10] which suggested that the most remarkable energy retention was obtained in cattle fed a diet that improved digestibility, decreasing energy lost through feces and enteric methane. Our results suggest that strategic feeding using cassava can enhance nutrient and energy supplies for Holstein young bulls.

The pooled data analysis in this study resulted in a metabolizable energy requirement for maintenance of 599 kJ/kg BW^0.75 and the efficiency of ME utilization for growth was 0.88 for Holstein crossbreed young bulls. This value is greater than recommended for Bos taurus [12,23]. Our values are also higher than the published ME_m values (458 to 541 kJ/kg BW^0.75) reported for Zebu cattle [2,4,8,10]. This study confirmed that the energy requirement for maintenance in dairy crossbreed cattle (B. indicus × B. taurus) is greater than in Zebu cattle (B. indicus).

5. Conclusions

Replacing cassava with rice straw from 5% to 35% in Holstein young bull’s diets substantially reduced physically effective fiber, chewing, and ruminating time but improved nutrient and net energy intake. Additionally, rumen fermentation characteristics shifted and enhanced by increasing total volatile fatty acid and propionic acid concentrations and reducing enteric methane emissions, thus leading to a greater energy balance status because of the increase in daily feed intake and nutrient digestibility. We conclude that cassava is a good energy feed source for Holstein crossbred cattle.